* Some trial and error as I try to power Seeduino Ethernet and 24 x Grove Chainable RGB LED *
By default the grove cables supply power to the LEDs from the seeduino. However this is too much current for the seeduino and it gets very hot.
So we need to divert the power lines for the LEDs to another source but keep the data lines going to Seeduino.
The problem is that the data lines and power lines use the same ground wire.
This seems like an easy solve right? Just bridge the ground lines between the Seeduino Ethernet power source and LED power source right? Like Adafruit have done on their tutorial for their very similar P9813 LED strip?
In our case this won’t work because the power supplies are different voltages and so can’t be the same supply with the same ground.
The Seeduino Ethernet needs 6 – 12v on it’s jack, the RGB LEDs need 5v.
Test A: I’m going to try a simple hack, which is powering the Seeduino Ethertnet via it’s FTDI connection at 5v directly. Then it will be ok to share the ground of these.
It’s not documented that you can do this. Neither the Seeduino Ethernet documentation or ArduinoBoardEthernet documentation mention powering via FTDI. So hopefully it doesn’t blow it up.
– Results 1
Getting warm, but not hot.
Ran for 10 minutes then froze when I bumped ethernet connection. Doesn’t seem to recover from having ethernet fail without a hard reset.
– Results 2
Ran for 30 minutes before I screwed it up by disconnecting my network.
– Results 3
Now connected to network wirelessly via a TPLink-TLWR702N in client mode.
Ran for 1hr:30 before I accidentally reset it.
Still kind of warm.
– Result 4
Ran for 30 mins before I accidentally reset it again.
This seems to be happening pretty easily, going to try an ethernet cable with a locking plug.
– Result 5
Better phsyical reliability with a locking plug rj45. Shook it around a whole bunch and no freezes, still giving good data. Guess I had a particularly bad connector on my previous cable. Still, it would be better if the seeduino could withstand losing it’s network connection for a moment without freezing.
LEARNING THE HARD WAY: Seems the disconnect issue with the Seeduino is only a problem when the rj45 is pulled from the seeduino itself. Pulling it at the destination or the network being unavailable in other ways doesn’t seem to bother it. So far in any case, needs more testing.
=== Conclusion ===
– You can power the Seeduino Ethernet using a regulated 5v source via the 5V and gnd pins on the FTDI header. However it gets quite warm, I would have reservations about using it for an ‘always on’ application.
– Large groups of the grove Chainable RGB LED can be powered by directly powering the chain post-microcontroller, as long as you share the ground connection between the microcontroller and seeduino.
– The seeduino ethernet network functionality freezes if you disconnect the cable while it’s working on UDP (and perhaps in other situations). The only way to get it back is a hard reset.
Reliability Test: Seeduino Ethernet UDP + Grove Accelerometer(±16g) ADXL345r + Grove Chainable RGB LED
* A log of reliability tests and subsequent findings when working with the Seeduino Ethernet, Grove ADXL345 Accelerometer and Grove Chainable RGB LEDs * Read more…
* A log of reliability tests and subsequent findings when working with the Uno R3, Grove Accelerometer ADXL345 and Grove Chainable RGB LED * Read more…
* A log of reliability tests and subsequent findings when working with the Uno R3, Grove Accelerometer MMA7660FC and Grove Chainable RGB LED * Read more…